A quantitative approach to predict tunnel overbreak based on the Q-system

Overbreak can be a significant challenge in tunnelling and is usually assessed on a project-by-project basis. This paper attempts to quantify tunnel overbreak in jointed rock masses by producing a predictive model that practitioners can utilize at the preliminary design stage. Discontinuum numerical analysis of a circular-shaped tunnel is performed using the combined finite difference – discrete element method to generate a database of rock masses prone to overbreak. The Q-system, rock mass quality index, is used to classify the rock masses that are expected to overbreak in various tunnelling environments. The results are scrutinized to identify trends between overbreak and: i) Q, ii) average joint spacing and, iii) in-situ stress conditions. It is observed that overbreak increases when Q < 4 and is uniform when Q ≤ 4. A predictive model is generated through multiple regression analysis and can be used to identify and highlight the maximum likely overbreak or scenarios where overbreak is expected to occur.